• Journal of Intelligence and Information Systems
• /
• v.21 no.1
• /
• pp.29-45
• /
• 2015

Response modeling is a well-known research issue for those who have tried to get more superior performance in the capability of predicting the customers' response for the marketing promotion. The response model for customers would reduce the marketing cost by identifying prospective customers from very large customer database and predicting the purchasing intention of the selected customers while the promotion which is derived from an undifferentiated marketing strategy results in unnecessary cost. In addition, the big data environment has accelerated developing the response model with data mining techniques such as CBR, neural networks and support vector machines. And CBR is one of the most major tools in business because it is known as simple and robust to apply to the response model. However, CBR is an attractive data mining technique for data mining applications in business even though it hasn't shown high performance compared to other machine learning techniques. Thus many studies have tried to improve CBR and utilized in business data mining with the enhanced algorithms or the support of other techniques such as genetic algorithm, decision tree and AHP (Analytic Process Hierarchy). Ahn and Kim(2008) utilized logit, neural networks, CBR to predict that which customers would purchase the items promoted by marketing department and tried to optimized the number of k for k-nearest neighbor with genetic algorithm for the purpose of improving the performance of the integrated model. Hong and Park(2009) noted that the integrated approach with CBR for logit, neural networks, and Support Vector Machine (SVM) showed more improved prediction ability for response of customers to marketing promotion than each data mining models such as logit, neural networks, and SVM. This paper presented an approach to predict customers' response of marketing promotion with Case Based Reasoning. The proposed model was developed by applying different weights to each feature. We deployed logit model with a database including the promotion and the purchasing data of bath soap. After that, the coefficients were used to give different weights of CBR. We analyzed the performance of proposed weighted CBR based model compared to neural networks and pure CBR based model empirically and found that the proposed weighted CBR based model showed more superior performance than pure CBR model. Imbalanced data is a common problem to build data mining model to classify a class with real data such as bankruptcy prediction, intrusion detection, fraud detection, churn management, and response modeling. Imbalanced data means that the number of instance in one class is remarkably small or large compared to the number of instance in other classes. The classification model such as response modeling has a lot of trouble to recognize the pattern from data through learning because the model tends to ignore a small number of classes while classifying a large number of classes correctly. To resolve the problem caused from imbalanced data distribution, sampling method is one of the most representative approach. The sampling method could be categorized to under sampling and over sampling. However, CBR is not sensitive to data distribution because it doesn't learn from data unlike machine learning algorithm. In this study, we investigated the robustness of our proposed model while changing the ratio of response customers and nonresponse customers to the promotion program because the response customers for the suggested promotion is always a small part of nonresponse customers in the real world. We simulated the proposed model 100 times to validate the robustness with different ratio of response customers to response customers under the imbalanced data distribution. Finally, we found that our proposed CBR based model showed superior performance than compared models under the imbalanced data sets. Our study is expected to improve the performance of response model for the promotion program with CBR under imbalanced data distribution in the real world.

• Journal of Korean Society of Forest Science
• /
• v.101 no.1
• /
• pp.7-19
• /
• 2012

After having announced legislation for threatened or endangered species on the List of Endangered and Threatened Wildlife and Plants in 2005, the Korea Ministry of Environment proposed (in June 2011) amending the list, thereby delisting or reclassifying endangered species using new quantitative criteria for two levels (I and II), as well as status reviews. The new legislation included 40 species remained in their original endangered status, but 19 species were delisted, 5 species were proposed as candidates for delisting, 29 species were given a new endangered listing, and 3 species were proposed for an endangered listing in Korea. We assessed the threatened status of 98 plants using the IUCN Red List Criteria (version 3.1) at the global level, and compared the Ministry's revised criteria with the IUCN Red List Criteria and ESA criteria used in the USA. Most species proposed by the Ministry do not qualify as threatened and one of the major difficulties found in applying IUCN Red List Criteria at the global scale was a lack of knowledge on the status of species at broader geographic scales and the perceived difficulty this causes. Under the current classification process, many endangered species, such as Abeliophyllum distichum, Leontice microrhyncha, Echinosophora koreensis, Leontopodium coreanum, Iris odaesanensis, and Corylopsis coreana at global level were excluded here. Knowledge gaps and uncertainties mean that the number of taxa at high risk of extinction may be substantially greater than is currently understood. Due to a lack of information on its taxonomic status, currently there is controversy over the Red List status of Physocarpus insularis. Also, Caragana koreana, which was an invalidly published name, should be excluded here. Although the Korea Ministry of Environment insisted this procedure was conducted by applying the modified IUCN threat categories and definitions, this evaluation has been carried out based only on subjective views and misapplication of the IUCN Red List Criteria. The current listings by the Korea Ministry of Environment should be challenged. We suggest that broad species concepts on endemic species are applied and also criteria that adequately address the proper quantitative knowledge should be used. It is suggested that the highest priorities for the Red List should be given to endemic species at least in the Korean peninsula first at global scale.

• Korean Journal of Heritage: History & Science
• /
• v.53 no.2
• /
• pp.222-241
• /
• 2020

The Taereung Royal Tomb from the Joseon Dynasty is the tomb of Empress Munjeong, the second queen of King Jungjong, and it contains various types of stone artifacts. All of these stone artifacts were constructed using coarse- to medium-grained biotite granite. The major types of deterioration of the stone artifacts are identified as surface weathering and biological contaminants. Exfoliation (145 sculptures), granular decomposition (138 sculptures), and repair materials (156 sculptures), along with biological contaminant algae (154 sculptures), lichen (165 sculptures) and moss (97 sculptures), have a high occurrence frequency. In particular, it is deemed that immediate conservation treatment is required, as biological deterioration (algae) represents the most serious condition (grade 3 or higher in 94% of all stones), and it is thought that exfoliation and granulation decomposition are required for long-term conservation management. As a result of equo -tip hardness and ultrasonic measurement, more than 70% of stones were found to have very weak physical properties. Through hyperspectral analysis, organisms were shown to inhabit more than 80% of the surface of burial mound stone artifacts, and P (phosphorus), S (sulfur), Cl (chlorine), and Ca (calcium) were detected in this area. This is because Taereung Royal Tomb has been exposed to the outdoors for hundreds of years and has been weathered by physical, chemical, and biological factors. Therefore, among the stone artifacts in the Taereung Royal Tomb, those with high physical weathering grades are considered to require consolidation to reinforce them physically. Since organisms are highly likely to cause stone damage, they must be removed via dry and wet cleaning. In addition, in order to delay the reoccurrence of organisms following conservation treatment, it is necessary to regularly clean up the soil that has flowed into the burial mound, and to monitor conservation conditions over the long term.

• Journal of Chest Surgery
• /
• v.39 no.3 s.260
• /
• pp.220-225
• /
• 2006

Background: Spontaneous pneumomediastinum is an uncommon, benign, self-limited disorders that usually occurs in young adults without any apparent precipitating factors or disease. The purpose of this study was to review our experience in dealing with this entity and describe a reasonable course of assessment and management. Material and Method: A retrospective case series was conducted to identify adults patients with SPM who were diagnosed and treated in a single institution between 2001 and 2005. Result: Fifteen patients were identified who included 14 men and 1 women with a mean age of 26 years. Presenting symptoms were chest pain in 12 patients ($80\%$), dyspnea in 5 patients ($33\%$), and throat discomfort in 4 patients ($26\%$). Two cases were associated with use of inhalational drugs and 3 cases were associated with exercise. The predisposing factors were asthma, excessive exercise, and vomiting in spontaneous pneumomediastinum. The physical findings were subcutaneous emphysema in 10 patients ($77\%$). Chest radiography and computerized tomography were the diagnostic methods in all cases with CT scan revealing six cases with associated pulmonary abnormalities. Esophagogram and flexible bronchoscopy were selectively used. Fifteen patients ($100\%$) were admitted to the hospital. Their mean hospital stay was 3 days. All patients were conservatively treated. In a follow-up of 3 years no complications or recurrences were observed. Conclusion: Most simple spontaneous pneumomediastinum cases were benign diseases and most of them ($77\%$) had shown typical chest pain, dyspnea and subcutaneous emphysema. Inhalational drug use was not a major cause of SPM; however, increased use of bronchoinhalers was a suspicious cause of SPM.

• Neonatal Medicine
• /
• v.15 no.1
• /
• pp.80-83
• /
• 2008

Purpose : Inguinal hernias are common in children and sometimes are associated with dangerous complications, such as incarceration. There are no established management guidelines for ovarian hernias. We have reviewed the clinical course of ovarian hernias in infants. Methods : We reviewed the medical records of female infants diagnosed with ovarian hernias by ultrasonogram at Kwandong University College of Medicine, Cheil General Hospital, and the Women's Healthcare Center between March 2001 and August 2007. We analyzed the patients gestational age, birth weight, age at the time of detection of the inguinal mass, the patients chief complaints, operative time, post-operative complications, and ultrasonographic findings. Results : Eight female infants had ovarian hernias, four of whom were born prematurely. Seven infants had left-sided ovarian hernias, and one infant had a right-sided ovarian hernia. Five infants underwent surgery and there were no postoperative complications or recurrences. Three girls did not have surgery, and the ovarian hernias regressed spontaneously, with no recurrences or complications. The regression time of inguinal masses ranged from 70-161 days after birth. Conclusion : Physical examination to detect movable masses within the labium major in premature female infants is important because the incidence of premature inguinal hernias is much higher than in term infants. No rational medical treatment plans for female ovarian hernias have been published to date. We cared for three girls with spontaneous regression of ovarian hernias. Pediatricians should be aware whether emergent surgery for ovarian hernias is indicated.

• Korean Journal of Ecology and Environment
• /
• v.36 no.2 s.103
• /
• pp.181-190
• /
• 2003

An investigation was conducted on the aquatic environment of the Okchon Stream watershed six times from May to September 2002. The results of investigation revealed that variation of environmental factors were quite significant for each stream and reach, showing a significant difference between running water and stagnant water. Aquatic nutrients were relatively low in the upstream, gradually increasing as the influx of treated wastewater into the stream increased. This suggests that the point source definitely affected the nutrient content of the stream. In particular, the variations of SRP and $NH_4$ were very distinct in the watershed compared to other nutrients. Thus, it can be considered as a major factor in evaluating the effect of treated wastewater. Immediately after the influx of treated waste-water, the average content of SRP rose to 919.3 ${\mu}g$ P/l. This was a very effective level in the watershed, suggesting that the percentage of the nutrients in the water was controlled by the content of P. The constant supply of treated wastewater was found to be a critical factor in triggering the increase in chl-a in the embayment of the stream. With the proliferation of the blue-green algae, the content of chl- a ranged 234.5${\sim}$1,692.2 ${\mu}g/l$. The maximum standing crops exceeded $1.0{\times}10^6$ cells/ml in August, which was more than 200 times the level for red tide in the freshwater. This result was well reflected in other environmental factors, with 100% of AFDM/TSS reflecting the severity of water pollution by algae. Therefore, the reduction of P and N con-tents in the treated wastewater is critical in improving the aquatic environment of the stream as well as water quality management for the reservoir.

• Korean Journal of Ecology and Environment
• /
• v.49 no.4
• /
• pp.354-374
• /
• 2016

This study explored spatiotemporal variability of water quality in correspondence with hydrometeorological factors in the five stations of Paldang Reservoir located in the Han River during 4 years from May 2012 to December 2015. Variability of basic water quality factors were largely related with seasonal fluctuations of hydrology. Temperature stratification occurred in the deep dam station, and prolonged hypoxia was observed during the draught year. Nitrogen nutrients were increased with decreasing inflow in which changing pattern of $NH_4$ reversed to $NO_3$ by the effect of treated wastewater effluent. Phosphorus increase was manifest during the period of high inflow or severe drought. Chl-a variation was reversely related with both flow change and AGP(algal growth potential) variations. Our study demonstrated that water quality variability in Paldang Reservoir was largely attributed to both natural and operational changes of inflow and outflow (including water intake) based on major pollution source of the treated wastewater (total amount of $472{\times}10^3m^3d^{-1}$) entering to the water system from watershed. In the process of water quality variability, meteorological (e.g., flood, typhoon, abnormal rainfall, scorching heat of summer) and hydrological factors (inflow and discharge) were likely to work dynamically with nutrients pulse, dilution, absorption, concentration and sedimentation. We underline comprehensive limnological study related to hydro-meteorolology to understand short- and long-term water quality variability in river-type large reservoir and suggest the necessity of P-free wastewater treatment for the effective measure of reducing pollution level of Paldang drinking water resource.

• Korean Journal of Environment and Ecology
• /
• v.33 no.5
• /
• pp.497-505
• /
• 2019

Winter hibernation in wildlife is a unique physiological mechanism for survival. For Asiatic black bears (ABBs, Ursus thibetanus ussuricus), hibernation is a very important but dangerous time for the cubs to be born. This study surveyed ABBs living in Mt. Jiri To examine the relationship between the temperature during hibernation and the hibernation days. The survey found that the average start and end dates of hibernation was December 7 and April 20, respectively. The average day of hibernation for females who gave birth was $167.8{\pm}22.6$ days. The average temperature of 5 days before hibernation was $0.6{\pm}4.1^{\circ}C$, the average temperature during hibernation was $1.3{\pm}2.43^{\circ}C$, and the average temperature of five days before the end of hibernation was $12.6{\pm}3.1^{\circ}C$. The hibernation days of the females that did not give birth to cubs and the males were $120.4{\pm}25.7$ days and $113.6{\pm}25.8$ days, respectively. The average temperatures of 5 days before hibernation was $2.1{\pm}4.2^{\circ}C$ and $-1.8{\pm}3.9^{\circ}C$, respectively. The average temperature during hibernation was $-0.2{\pm}1.6^{\circ}C$ and $-0.4{\pm}2.4^{\circ}C$. The average temperatures of five days before the end of hibernation were $7.8{\pm}3.6^{\circ}C$ and $7.8{\pm}4.4^{\circ}C$. These results indicate that females giving birth to cubs have higher hibernation days and average temperatures than the females that do not give birth and the males, which is due to the process of raising cubs during hibernation. The hibernating days and mean temperature for the groups in each lifecycle did not show any difference between the groups. This study is meaningful in that it disclosed the characteristics of hibernating intrinsic behaviors of ABBs. It observed the specific hibernation period and temperature of ABBs bears inhabiting in Mt. Jiri and examined the difference by sex, female(giving birth) and life cycle group according to temperature. The results of this study can be used to prevent the conflict between ABBs and humans in winter and spring and establish the preservation management plan.

• The Journal of the Petrological Society of Korea
• /
• v.25 no.3
• /
• pp.169-193
• /
• 2016

In order to characterize the Neotectonic crustal deformation and current stress field in and around the Korean Peninsula and to interpret their tectonic implications, this paper synthetically analyzes the previous Quaternary fault and focal mechanism solution data and recent geotechnical in-situ stress data and examines the characteristics of crustal deformations and tectonic settings in and around East Asia after the Miocene. Most of the Quaternary fault outcrops in SE Korea occur along major inherited fault zones and show a NS-striking top-to-the-west thrust geometry, indicating that the faults were produced by local reactivation of appropriately oriented preexisting weaknesses under EW-trending pure compressional stress field. The focal mechanism solutions in and around the Korean Peninsula disclose that strike-slip faulting containing some reverse-slip component and reverse-slip faulting are significantly dominant on land and in sea area, respectively. The P-axes are horizontally clustered in ENE-WSW direction, whereas the T-axes are girdle-distributed in NNW direction. The geotechnical in-situ stress data in South Korea also indicate the ENE-trending maximum horizontal stress. The current crustal deformation in the Korean Peninsula is thus characterized by crustal contraction under regional ENE-WSW or E-W compression stress field. Based on the regional stress trajectories in and around East Asia, the current stress regime is interpreted to have resulted from the cooperation of westward shallow subduction of the Pacific Plate and collision of Indian and Eurasian continents, whereas the Philippine Sea plate have not a decisive effect on the stress-regime in the Korean Peninsula due to its high-angle subduction that resulted in dominant crust extension of the back-arc region. It is also interpreted that the Neotectonic crustal deformation and present-day tectonic setting of East Asia commenced with the change of the Pacific Plate motion during 5~3.2 Ma.

• Korean Journal of Soil Science and Fertilizer
• /
• v.43 no.6
• /
• pp.789-792
• /
• 2010

Major impacts of climate change expert that soil erosion rate may increase during the $21^{st}$ century. This study was conducted to assess the potential impacts of climate change on soil erosion by water in Korea. The soil loss was estimated for regions with the potential risk of soil erosion on a national scale. For computation, Universal Soil Loss Equation (USLE) with rainfall and runoff erosivity factors (R), cover management factors (C), support practice factors (P) and revised USLE with soil erodibility factors (K) and topographic factors (LS) were used. RUSLE, the revised version of USLE, was modified for Korean conditions and re-evaluate to estimate the national-scale of soil loss based on the digital soil maps for Korea. The change of precipitation for 2010 to 2090s were predicted under A1B scenarios made by National Institute of Meteorological Research in Korea. Future soil loss was predicted based on a change of R factor. As results, the predicted precipitations were increased by 6.7% for 2010 to 2030s, 9.5% for 2040 to 2060s and 190% for 2070 to 2090s, respectively. The total soil loss from uplands in 2005 was estimated approximately $28{\times}10^6$ ton. Total soil losses were estimated as $31{\times}10^6$ ton in 2010 to 2030s, $31{\times}10^6$ ton in 2040 to 2060s and $33{\times}10^6$ ton in 2070 to 2090s, respectively. As precipitation increased by 17% in the end of $21^{st}$ century, the total soil loss was increased by 12.9%. Overall, these results emphasize the significance of precipitation. However, it should be noted that when precipitation becomes insignificant, the results may turn out to be complex due to the large interaction among plant biomass, runoff and erosion. This may cause increase or decrease the overall erosion.